论文部分内容阅读
降雨资料时间序列长度是计算多年平均降雨侵蚀力过程中的重要不确定性因素。论文以中国601个气象站1980—2009年逐月降雨资料为数据源,利用Wischmeier经验公式计算了各气象站逐年降雨侵蚀力(R因子),用简单随机抽样方法抽取样本容量分别为30 a、20 a、10 a和5 a四种不同的R值样本,计算了R平均值相对允许误差10%和25%条件下抽样估计的置信度。结果表明:降雨资料的时间序列长度对R平均值的估计置信度有显著影响;R平均值置信度存在明显的地域差异,长江以南、青藏高原东部以及河西走廊南部的祁连山地区置信度较高;在降雨资料有限的情况下,必须根据土壤侵蚀研究的精度要求分析R平均值的抽样误差及其置信度,以保证土壤侵蚀定量预报的客观性与准确性。
The length of time series of rainfall data is an important uncertainty factor in the process of calculating the mean annual rainfall erosivity. In this paper, monthly rainfall data of 601 weather stations in China from 1980 to 2009 are used as data sources, and annual rainfall erosivity (R factor) of each weather station is calculated by using Wischmeier empirical formula. The sample sizes extracted by simple random sampling method are 30 a, 20 a, 10 a and 5 a, four different R-value samples were used to calculate the confidence of the sample estimate under the relative allowable error of 10% and 25% of R average. The results show that the length of time series of rainfall data has a significant influence on the estimated confidence of R average. There are obvious regional differences in the confidence of R average value. The confidence of Qilian Mountain region to the south of the Yangtze River, the eastern Qinghai-Tibet Plateau and the southern Hexi Corridor is high In the case of limited rainfall data, the sampling error and its confidence of R average must be analyzed according to the accuracy requirements of soil erosion research so as to ensure the objectivity and accuracy of quantitative prediction of soil erosion.